We need to remember that solar capacity has long life-span (say ~25 years minimum) per panel. Where coal needs to be constantly mined and burned.
If you multiple 0.087*25 years of operation you get 2.175GWh/worker in solar.
Now, there are an almost endless numbers of adjustments you can make to this calculation. And you really need to use the "newly installed capacity/year" number when multiplying by the lifespan of the panels.
(These people say the US installed 4.8GW in 2013: http://cleantechnica.com/2014/03/18/37-gw-solar-capacity-ins... I have really no idea if that's accurate)
I think the calculation is more complicated than you make it seem. And it's not surprising that the cost to install a long-life span technology is relatively high (in terms of man-power) as solar is in its infancy.
(I say infancy b/c I believe we'll be far more efficient at making and installing solar cells 30 years from now than we are today. I expect coal mining to make smaller efficiency gains.)
Finally, jobs are a great thing. YC put out a RFS looking for ~1million new jobs: https://www.ycombinator.com/rfs/#million I'd love to see them created in solar.
Panel lifetime is generally given as a nominal 20-25 years. That's not a minimum. The reason is a gradual degradation, mean of 0.8%, median 0.5% of output per year, due to multiple causes. See: "Photovoltaic Degradation Rates: An Analytical Review", Dirk C. Jordan and Sarah R. Kurtz, NREL
http://www.nrel.gov/docs/fy12osti/51664.pdf
I would like to know how the employment numbers here are calculated, and how they're allocated across the lifetime of solar plant. I'm not trying to knock solar, I'm showing that it's quite possible to have too much of a good thing though.
CleanTechnica is generally a pretty reliable source. US EIA gives 3.2 GW nameplate capacity for solar as of 2012: http://www.eia.gov/electricity/annual/html/epa_04_03.html Installations have been going at a major clip, so 4.8 GW sounds highly plausible.
"I think the calculation is more complicated than you make it seem."
It almost certainly is. However my point was that it's also a bit less straightforward than the BusinessInsider piece presents. Believe me, I'd like to see good jobs in the solar sector. But I'd really like for those to be productive jobs.
I do think that solar will continue to see cost reductions (and those will be key to its success). Addressing factors other than just efficiency (mostly overrated), and in particular, the installation, durability, and maintenance costs, are what I suspect are key.
Jobs ... are complicated. My point isn't so much that they're good or bad, more that if the trends reported are projected forward, we'll see a very large portion of the labor force directly employed in energy harvesting. This isn't the end of the world, but it does mean that roughly 1 in 5 present jobs will be supersceded. There was an earlier time when energy harvesting occupied a similar or larger portion of the workforce, though it was called "farming" at the time. In some countries it still comprises 90% or more of all work.
This plays into another discussion -- that of EROEI and what levels of it are required to support various levels of economic and technological complexity.
http://www.rmi.org/cms/Download.aspx?id=11218&file=2014-11_R...
Australia and Germany have lower costs per KW installed, but the installation cost exceeds panel cost even there. This is a big problem. Even if panel cost goes to 0, installed cost will be maybe half what it is now.
Solar panels which are also roof components have been suggested for new construction, but that requires more roof design standardization than the housing industry is used to.
There are also "off-roof electrical costs". A solar installation usually requires a new meter box and additional boxes for inverters, control gear, and disconnects. Standardizing how solar systems connect at the meter box would simplify that.
The norm in Australia is now a one-day install. The US and Germany run slower, partly because more waterproofing is required. ("Every hole in the roof is a dagger through the heart of a house")
"Even if panel cost goes to 0, installed cost will be maybe half what it is now."
I've drawn the relationship between this and Amdahl's law, in parallel computing. As parallelization increases, it's the serial component of processing which comes to dominate. Eventually you reach a point where further parallelization yields no benefit.
I'm curious about your username. I know of animats from Carl Zimmer's article on them a ways back, and found the concepts of complexity and capabilities in evolution fascinating.
I suspect many of the jobs are in sales and marketing as well, which may well get optimized out.
But it should be added,
a) if we did this, you can bet on vastly improved automation and efficiencies, and much larger plants. Solar jobs are vastly dominated by rooftop installation, which isn't at all what you'd get if you were powering a whole nation.
b) After the solar output is built, you don't need to add much more, and maintenance is much less than the construction labor. You might need only 2 - 10 % of your estimate -- just adding solar at the electric grid replacement rate. Still, that's a lot of labor.
Once built, solar will need replacement at ~5% of installed base per year (based on a 20% lifespan).
Yes, large-installation construction should be less expensive than rooftop retrofitting. I also suspect we'll see increasing amounts of integrated solar construction.
As I note elsewhere, I haven't seen a good allocation of labor by phase of solar deployment (e.g., new rollouts, maintenance, ongoing replacement, support and administrative positions, recycling).
Oh, and at some point, recycling of decommissioned panels will also be a thing.
High labor costs per unit of energy is a good thing. Coal mining is resource intensive, and a significant portion of your energy dollar is going into paying for the cost of mining and transport. If a kWh of solar costs the same as a kWh of coal, all things equal, it is better for society for that cost to be spread among many well paid, middle class laborers.
The fact that fossil fuels allow for a huge avoided cost (that of the prehistoric accumulation of sunlight and transformation into fuels) doesn't make low labor costs inherently bad.
I'm wrapping my head around various bits of the labor / wage equity question, but on balance I suspect it's largely independent of cost inputs to basic resources, including energy, which would mean that you're confounding two unrelated issues.
The goal should be to have sufficient accessible well-paying jobs. Not for them to necessarily be in low-productivity, but nonetheless essential, energy harvesting positions.
Though that may turn out to be the case.
8.968GWh of "sales" , but how much of that is spent mining it? And cleaning up after it w/ the pollution from burning, the damage to the environment etc etc?
Same story for Solar too, i know there is impact, but I would venture that the "costs" are much less, so to speak.
The proper metric is net delivered energy. Generation is a reasonable proxy for that, though EROEI would give you a better measure.
The EROEI of coal is typically very high -- 1:100 to 1:200 or better, vs. 1:6 to 1:20 for solar PV (estimates vary, and are dependent on specifics).
Regardless, if you're looking at delivered energy and divide by labor, you end up with the values I've shown. I haven't seen the case presented this way previously and feel it helps illuminate the situation.
Very rapid growth from a small base is not the sort of behavior that's likely to continue long, but at that rate, of growth, solar power would be a majority component of U.S. generating capacity in only 8 years.
I suspect it'll take somewhat longer than that, but there's no question the sector's expanding tremendously.
The US EIA (Energy Information Administration), part of the Department of Energy, regularly posts forecasts. http://www.eia.gov/
There is research published by the National Research Labs (also part of DOE), particularly the National Renewable Energy Lab in Colorado, NREL. See their Renewable Electricity Futures Study: http://www.nrel.gov/analysis/re_futures/
The Electric Power Research Institute, EPRI, an industry research organization. http://epri.com/
Various large utilities, particularly in the Western US, post forecasts. See especially PG&E and Southern California Edison.
Rocky Mountain Institute (RMI) are among the private research organizations looking at renewables and future projections. http://www.rmi.org/
WorldWatch Institute looks at many factors including energy: http://www.worldwatch.org/
Cleantechnica seems to be a pretty good online renewable energy source, without much of the fluff that infects the sector.
Internationally, the International Energy Agency (IEA), as well as various government organizations, particularly the European Commission. CSIRO is an Australian research organization comparable to the US National Labs.
The International Panel on Climate Change (IPCC) has an impressively comprehensive report on renewable energy technologies: http://srren.ipcc-wg3.de/report/
There's the REN21 report: http://www.ren21.net/gsr
The Fraunhofer Institute has extensive information on Germany's renewables growth ("Energiewende"): http://www.ise.fraunhofer.de/en/
My point is that the growth rates are spectacular, and as I recall have been markedly exceeding earlier projections.
These and other sources are referenced on the RenewableTech FAQ: http://www.reddit.com/r/RenewableTech/wiki/techfaq#wiki_reso...
On the negative side: the fraction of electricity generation provided by solar vs. coal is a very small fraction, meaning that you're looking at far more workers per GWh of actual energy production.
The 8 month total generation for 2012 includes 12,346 GWh of solar PV, vs. 1,105,161 GWh for coal generation.
http://www.eia.gov/totalenergy/data/monthly/pdf/sec7_5.pdf
This means that a solar worker is responsible for 0.087 GWh of generation. A coal miner is responsible for 8.968 GWh of electricity generation.
At this rate, we'd require 32.11 million workers in solar to replace existing coal generation capacity. While that would be a strongly effective employment program, it would also be devoting a vastly larger portion of the United States' labor force to the task of energy provisioning.
That's a significant chunk of the 147.3 million total civilian workforce as of October, 2014, at 22%.
http://www.bls.gov/news.release/empsit.a.htm